Week Five:

Seems as though I will have to archive this too. But next week. I guess my main problem this week was reading in the xml files and getting their attributes. Nataliya seems to have solved this however, by going directly to the graph and ignoring the registry, while I had focused exclusively on the registry. I'm getting the oh-so-lovely null pointer exception rather frequently, but I know what's causing it, but am having trouble fixing it because it involves removing items from the graph at the same time that I am iterating through it, and it definitely does not like that. I'll get it sometime today though. I'm just as stubborn as it is!

Week Four:

Well, if you looked at my log, you know that this was a problemmatic week. The whole "completely destroying my program" thing really set me back. At this point, I'm not even curious to know what I did, but I'm still fuming about the fact that I had it right there in my grasp and it slipped right through.

I'm sure that if some Prefuse expert comes across this, they'll burst out laughing at my struggles. Well, I guess that's why they're the "expert" and I'm the "novice". But the whole adding nodes to a pre-existing graph issue is still really frustrating. I'm not sure if I'm misunderstanding the whole node vs. visual item relationship or what, but all I'm trying to do is create nodes, set edges and set the coordinates!!! (If you take pity upon me, give me an email).

I take a look at the forums for Prefuse, and sometimes I wonder what the background is for these posters. What's their degree? What's their specialty, experience, project? I guess I want to know these things so that maybe I won't feel like a complete incompetant. Hey! If these problems keep coming like this, I'll have to archive this too. Gives me incentive to not have them, right?

Week Three:

When expanding the demo program, I encountered my own version of the halting problem. While not so irritating as the "true" halting problem, I got stuck. How on earth do I get the graph to stop? Well, figured that one out. And how on earth do I get solid coordinates of nodes that continue to move? Figured that one out too. I'll post my semi-solutions on the algorithms page, and finally get something up there!

I think that that's enough problems for one week.

Week Two:

Hmmm...well, I guess I sort of forgot about this section for week two. Let's see...Oh, k-means! Yeah, that was a big problem implementing the k-means algorithm that Jung had set up, mostly because it was using some data structures that I had never used before. Believe it or not, I had never used HashMaps before, in C++ or Java! Strange, huh? So I had to figure out all of the ins and outs of HashMaps, Collections, Sets, etc... But it all worked out in the end.

Week One:

The only problem that I've run into so far has been a lack of research papers available on the web. And Bryn Mawr's access to scientific journals is deplorable. Nothing is full text, and the abstracts are just that: abstract. Have you noticed that computer science books are ridiculously expensive? Even if I wanted to get some background reading, I'd have to shell out about $150 per book! Forget it! What I don't understand is this lack of online materials. You'd think that computer science journals would be some of the first on the web, but no. Well, research now shouldn't be as intensive as it was in the past week, so my frustration should be down to a minimum.